[Gmsh] Problems with sceond order tetrahedral mesh

[longonig at comcast.net]

Dr. Geuzaine, 

I am using GMSH to mesh simple shapes using 2nd order tetrahedral elements (10-nodes). I noticed that GMSH outputs an error at the end of the 3D meshing process: negative jacobians. My FEM code also has difficulties in calculating the Jacobian for these elements. I found out in the online documentation that the order of the mid-side nodes 7-8-9 may have to be inverted as 7-9-8, this way maintaining the same order as the other nodes. If I do this my code calculates the Jacobian, but the results are wrong. What is also puzzling is the fact that I cannot even visualize the mesh in VTK format. Meaning, the mesh comes out with holes. Have you encountered this type of issue before? 



Regards, 





Gianluca Longoni, Ph.D. 
Website: http://plaza.ufl.edu/longoni/ 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20120921/0d519c88/attachment.html>